mcp-goodnews

MCP Goodnews is a simple Model Context Protocol (MCP) application that features a server for getting good, positive, and uplifting news. This tool fetches news articles from the NewsAPI and uses a Cohere LLM to rank and return the top news articles based on positive sentiment.

Read the blog post on Medium!

In a world where negative news often dominates headlines, Goodnews MCP aims to shine a light on more positive and uplifting news stories. This project was inspired by an earlier initiative called GoodnewsFirst, which delivered positive news daily to email subscribers — it was a really awesome project! While GoodnewsFirst predated recent breakthroughs in Large Language Models (LLMs) and relied on traditional methods for sentiment ranking, Goodnews MCP leverages modern LLMs to perform sentiment analysis in a zero-shot setting.

• Cohere API Key
• NewsAPI Key
• Claude Desktop Application
• uv Python Project and Package Manager

# Clone the repository
git clone https://github.com/VectorInstitute/mcp-goodnews.git

In the next step, we'll need to provide the absolute path to the location of this cloned repository.

# Navigate to the configuration directory
cd ~/Library/Application\ Support/Claude/config

# Edit the claude_desktop_config.json file
nano claude_desktop_config.json

# Navigate to the configuration directory
cd %APPDATA%\Claude\config

# Edit the claude_desktop_config.json file
notepad claude_desktop_config.json

And you'll want to add an entry under mcpServers for Goodnews:

{
  "mcpServers": {
    "Goodnews": {
      "command": "<absolute-path-to-bin>/uv",
      "args": [
        "--directory",
        "<absolute-path-to-cloned-repo>/mcp-goodnews/src/mcp_goodnews",
        "run",
        "server.py"
      ],
      "env": {
        "NEWS_API_KEY": "<newsapi-api-key>",
        "COHERE_API_KEY": "<cohere-api-key>"
      }
    }
  }
}

Claude Desktop will use the updated config to build and run the mcp-goodnews server. If successful, you will see the hammer tool in the bottom-right corner of the chat dialogue window.

Clicking the hammer tool icon will bring up a modal that lists available MCP tools. You should see fetch_list_of_goodnews listed there.

Example prompts:

• "Show me some good news from today."
• "What positive things happened in the world this week?"
• "Give me uplifting news stories about science."

1. When you request good news, the application queries the NewsAPI for recent articles
2. The Cohere LLM analyzes the sentiment of each article
3. Articles are ranked based on positive sentiment score
4. The top-ranking good news stories are returned to you through Claude

Apache 2.0

Stay positive with Goodnews MCP!